Make ADAS technology more popular in vehicles

alan000345

Make ADAS technology more popular in vehicles [Copy link]

 

查看详情

Advanced driver assistance systems (ADAS) features have been proven to reduce accidents and save lives. According to the Insurance Institute for Highway Safety in Consumer Reports, cars equipped with forward collision warning and automatic emergency braking systems experienced 50 percent fewer front-to-rear crashes than cars without these systems in 2017. Unfortunately, most accidents occur in vehicles that don’t have even the most basic ADAS applications installed.

As ADAS continues to advance toward Level 4 and Level 5 autonomous vehicles as defined by the Society of Automotive Engineers, we have the opportunity to have a greater impact on the road by creating autonomous vehicle technology that can be used in a wider range of vehicles.

Although it is not economically feasible to equip all cars with full ADAS technology, the goal should still be to equip as many cars as possible with driver assistance features. This means that more vehicles on the road need to be able to efficiently sense, process and respond to real-time data.

The need for smart and diverse sensing

Traditionally, image data collected for ADAS operation is analyzed by function-based computer vision algorithms. Computer vision has served the industry well over the past decade, but as ADAS operations become more advanced, designers need additional tools to process and adapt to the situations that drivers and their vehicles face on the road.

Keeping ADAS functioning in different road conditions is a challenge. When encountering unexpected situations such as bad weather or poor road conditions, the vehicle needs to adapt in real time. These situations are difficult to handle with traditional models, but by developing a dynamic system that helps the car perceive, understand and react quickly to the world around it, the car itself can become a powerful co-pilot for the driver. Such a system requires data and the ability to process data in real time with a combination of computer vision and efficient deep learning neural networks.

ADAS solutions need to extract data from different sets of sensors and convert them into behavioral intelligence for the vehicle. These sensors need to be equipped with different types of cameras and related optical, radar and ultrasonic technologies at a minimum; in more complex cases, lidar and thermal night vision devices are also required. In addition, the system can locate the vehicle by comparing the features extracted from the sensor data with high-definition map data. The understanding and analysis of this multi-modal sensor data must be done in real time (new data arrives 60 times per second) without the need to set up a data center server in the back seat of the car.

Learn how to improve autonomous parking technology with Jacinto processors.

The solution must be road-ready

Just as drivers must take in multiple pieces of information simultaneously and make quick decisions to drive safely, so too must all ADAS applications, regardless of the level of autonomy. The importance of a high-performance system-on-chip (SoC) is that it can perform parallel processing without drastically cutting into power, temperature, component, and integration cost budgets. SoC solutions can scale from simpler cases (fewer sensors, lower resolution) to the most complex without compromising basic ADAS functionality or requiring system downgrades.

Application performance that can adapt to various types of vehicles is only one of the requirements. Systems must be developed cost-effectively to achieve widespread and effective use. The complexity of in-vehicle software is growing exponentially (today's code is as long as 150 million lines), which has led to a surge in development and maintenance costs. As systems become more aware of road conditions, their functional safety requirements will continue to change and evolve, and must meet strict automotive quality and reliability targets. It is these stringent requirements and realities that support and drive the development of the automotive electronics market.

The right SoC can address all of these needs. It can properly balance memory, I/O, and processing cores to meet system BOM targets based on a range of application requirements. The right SoC can also accommodate open software development methodologies, making it possible to reuse generated code multiple times and save effort in development and testing. The SoC can also be built with functional safety as a prerequisite from the beginning and have the necessary reliability and product life to enable automotive production lines to continue on the market for many years. As long as it is done well, it will be just around the corner to equip more cars with powerful ADAS functions (as shown in Figure 1).

查看详情

Figure 1: ADAS application example

How TI helps democratize ADAS technology

TI leveraged decades of automotive and functional safety expertise to design our Jacinto ^TM 7 processor platform, focusing on addressing sensing, parallel operation and system-level challenges.

We focus on aspects that have a significant impact on the entire system: combining excellent perception capabilities to monitor the car's surroundings in multiple directions with a car-centric design approach to optimize power and system cost.

The new Jacinto ^TM 7 processor series (including TDA4VM and DRA829V ) integrates key functional safety features on the chip, which can realize both safety-critical and non-safety-critical functions on one device; they also improve data management by combining high-speed and automotive interfaces. Jacinto ^TM 7 processors bring practical performance to automotive ADAS and gateway systems and help reduce system costs, thereby realizing the popularization and popularization of ADAS technology.

mei4

Very good knowledge points, thank you very much for sharing, thank you for your hard work, thumbs up!